US20090007050A1 - System for designing re-programmable digital hardware platforms - Google Patents

Abstract

A digital design system and method are provided for re-programmable hardware platforms, such as field programmable gate arrays (FPGAs) and other re-programmable system designs. The design system and method bridge the gap between what has previously been a development and prototyping platform used during the design phase of an electronic design system (EDS) project, and commercially viable re-programmable product platforms to replace non-programmable platforms, such as discrete processors and ASICs.

Description

BACKGROUND OF THE INVENTION
• 1. Field of the Invention
• The present invention relates generally to a system and method for designing digital circuits and, more particularly, to a system and method for designing re-programmable digital circuits.
• 2. Description of the Prior Art
• Until recently, it has been difficult to substitute programmable devices for non-programmable design approaches for a number of reasons. These reasons include the following:
• • 1. Programmable devices are more expensive on a per-unit basis than the dedicated circuitry they replace.
  • 2. Programmable devices operate more slowly than discrete circuitry.
  • 3. Programmable devices are inherently less power-efficient than discrete circuitry.
  • 4. System implementation on programmable hardware is inherently more difficult than on non-programmable platforms.
• At present, these factors have generally relegated programmable logic to three common applications:
• • 1. Using the programmable device as a development and prototyping platform during the design phase of a project.
  • 2. Using the programmable device for specialized, low-volume or high-value applications in which programmability and re-programmability benefits outweigh the cost or efficiency penalties imposed by their use.
  • 3. Using low-capacity and/or low-cost “commodity” programmable devices to replace a limited amount of the digital circuit, allowing custom, or so-called “glue logic,” to support conventional non-programmable discrete circuitry for a specific application.
• Typically, once the market size for a particular product reaches a certain level, developers have the option to move custom circuitry onto an Application Specific Integrated Circuit (ASIC). Implementing the bulk of a complex digital system onto a dedicated ASIC provides very high performance and efficiency at very low per-unit cost. However, as ASIC manufacturing has moved to ever smaller geometries, following the principle of Moore's Law, the design and tooling costs have risen to the level at which many applications cannot currently support this move to dedicated silicon.
• The same technology that has made ASIC design more difficult and expensive has benefited Field Programmable Gate Array (FPGA) design. Moving FPGAs to these smaller geometries (currently 90 nm) enables them to be smaller, faster, more efficient, and less expensive. The leaders in FPGA technology, Xilinx and Altera, are now competing aggressively in an emerging “commodity FPGA” marketplace, offering high-capacity FPGAs at prices that are a full order of magnitude lower than the previous generation.
• As a result, it is now possible to design complex digital systems that are hosted almost entirely inside a re-programmable device at a deliverable price. The potential benefits of designing on a re-programmable platform include:
• • 1. Shorter time-to-design with no delay while prototyping or tooling for mass production;
  • 2. Lower development cost with no tooling or prototyping costs;
  • 3. Lower development risk without need to scrap defective tooling and providing the ability to re-design on the fly; and
  • 4. In-field upgradability after a product is delivered.
• However, before these benefits can be realized, the problem of implementing complex hardware and software systems on a re-programmable platform must be addressed. While many engineers look to FPGA technology to provide higher levels of on-chip integration and a lower risk alternative to the cost and lead time of conventional ASICs, system-level design on an FPGA platform is a difficult exercise, particularly when it comes to bringing the processor into the FPGA.
• Presently, the design of such ambitious “system scale” designs has required the same level of design expertise as the design of ASICs. The skill, knowledge, and technology resources required for this level of design are largely restricted to large enterprises. For example, these designs are normally rendered in a highly cumbersome textual hardware description language (HDL). This language is an efficient vehicle for component, or micro-level, logic, but becomes unwieldy when applied at the systems level. Another problem is system verification, in which elaborate test strategies must be employed in order to simulate the operation of a complex system prior to its implementation.
• Thus, it would be desirable to provide a design system and method for re-programmable digital platforms which overcome the above limitations and disadvantages of conventional systems and techniques and facilitate the design of re-programmable digital platforms. It would also be desirable to provide re-programmable system design that is attainable for the broadest potential market, and to overcome a number of problems in re-programmable design implementation. It is to this end that the present invention is directed. Overcoming these obstacles has resulted in several novel solutions that have no precedent in the electronics design automation domain. The various embodiments of the present invention provide many advantages over conventional design systems and methods.
SUMMARY OF THE INVENTION
• One embodiment of the design system and method in accordance with the present invention provides many advantages over conventional design systems and techniques, which make the design system and method in accordance with the present invention more useful to digital circuit designers. The design system and method for re-programmable digital platforms in accordance with the present invention are based on a realization that the potential benefits of re-programmable hardware platforms, such as FPGAs, potentially outweigh the historic disincentives that have inhibited their wider adoption for digital systems.
• In accordance with the present invention, a digital design system and design process are provided for re-programmable hardware platforms, such as FPGAs and other re-programmable system designs. The design system and process bridge the gap between what has previously been a development and prototyping platform used during the design phase of an electronic design system (EDS) project, and commercially viable programmable product platforms to replace non-programmable platforms, such as discrete processors and ASICs.
• The approach employed by one embodiment of the design system and method for re-programmable digital platforms in accordance with the present invention is based on a process of translating the required design process back into models which are familiar and intuitively manageable by most electronics engineers. This allows users of the design system and method for re-programmable digital platforms in accordance with the present invention to design as if they are using discrete components, for example, processors and peripheral devices, in order to build complex systems with schematic symbology traditionally used with EDS. The design system and method for re-programmable digital platforms in accordance with the present invention require complex background automation in order to maintain the high-level abstraction of the design at the level of user interaction. Symbolic, component level logical blocks are provided to the user, such as a system design engineer. This generic logic can be quickly assembled into complex systems. The user does not need to write any RTL- or netlist-level descriptions, as these are managed automatically by the design environment. The pre-synthesis of these component-level blocks is enabled by a system that allows components to be symbolically rendered and interconnected within a supporting graphical design environment. The pre-synthesized components are supplied as object code entities without having to expose the underlying RTL- or netlist-level source code. The system includes a library having a comprehensive set of “components” that can be used by the user to add functionality to the design. These range from simple gate-level functional blocks to high-level hardware, such as multipliers and pulsewidth modulators, as well as high-level functions, such as microprocessors and communication functions. The design system and method for re-programmable digital platforms in accordance with the present invention manages the resources to instantiate the design in the target device.
• The design system and method for re-programmable digital platforms in accordance with one embodiment of the present invention automatically re-target a generic design for a selected programmable device transparently around each different tool chain and provide a generic view of the process flow to the user. This enables users to only have to learn one process flow in order to use multiple different vendors' tool chains, and allows easy movement between different digital devices from different vendors to easily port their designs to new target devices.
• One embodiment of the design system and method for re-programmable digital platforms in accordance with the present invention also automatically manages device-specific design constraints by providing a generic constraint model in which constraints can be specified. The design system and method for re-programmable digital platforms in accordance with the present invention transparently map these to the vendor-specific constraint system for implementation. For the minority of cases in which there are constraints that do not make sense for any other device, the design system and method for re-programmable digital platforms in accordance with the present invention allow these also to be specified in a generic way, but they then only map to specific devices.
• The design system and method for re-programmable digital platforms in accordance with one embodiment of the present invention employ a development platform for interactive, re-targetable hardware and software co-design of programmable device hosted systems by utilizing a device referred to as a NanoBoard to allow a single development platform to provide development resources for an unlimited number of target devices, such as FPGAs, CPLDs, or embedded microprocessors, from one or more different semiconductor manufacturers. The NanoBoard also contains a controller device that manages communications with the host-based development software, the daisy-chaining of multiple NanoBoards, connections to user-designed boards, and interfaces to physical resources on the NanoBoard. This device uses a so-called “NanoTalk” protocol to communicate with the PC-based software and to communicate with other NanoBoards in the daisy chain. Preferably, the clock frequency may be the controlled by the host-based development software. Each NanoBoard also contains one or more standard sockets for installing devices that may be used for development. These so-called daughter boards generate an identification code that is read by the controller which can then program the device at power-up. Preferably, the system for re-programmable digital platforms in accordance with one embodiment of the present invention comprises a programmable power supply for accommodating daughter boards with different power rail requirements.
• One embodiment of the design system and method for re-programmable digital platforms in accordance with the present invention provides an interactive design environment that automates support for multiple development boards. Each development board has a master and slave connector that allows an unlimited number of development boards to be daisy chained together. When another board is connected into the daisy chain, it is automatically recognized, and all of its resources become available to the system. This is controlled by a NanoTalk controller that detects the presence of the new board and then brings all of its hard devices into the main hard-chain, and all of its soft devices into the main soft-chain. The NanoTalk controller on each board can then be accessed from the PC-based software system to control the individual resources located on each development board. This system uses a single pin on a header to allow another board with JTAG devices to be automatically brought into the main JTAG chain. A protocol for multiplexed JTAG/SPI channels is used, in which the system allows a single interface between the PC and the development platform to control multiple JTAG and SPI chains. Other than the standard JTAG pins, a single mode pin is used to control the system.
• The design system and method for re-programmable digital platforms in accordance with one embodiment of the present invention preferably have a generic boot system capable of supporting multiple programmable device architectures, which allows a single controller and a generic low-cost serial flash RAM device to be used to program multiple different target devices. The daughter boards that are plugged into the system generate an identification code that is used by the controller to identify the device. At power-up, the controller identifies the target device, reads the data stream from the serial flash device, and programs the device using a programming algorithm that is appropriate for that target device.
• The design system and method for re-programmable digital platforms in accordance with the various embodiments of the present invention constitute a radical re-consideration of the digital system design process. The design system and method for re-programmable digital platforms in accordance with the present invention enable the design of FPGAs and other emerging re-programmable hardware architectures to ultimately replace non-programmable digital platforms, such as discrete processors and ASICs, as deliverable product platforms.
• The foregoing and other objects, features, and advantages of the present invention will become more readily apparent from the following detailed description of various embodiments, which proceeds with reference to the accompanying drawing.
BRIEF DESCRIPTION OF THE DRAWING
• The various embodiments of the present invention will be described in conjunction with the accompanying figures of the drawing to facilitate an understanding of the present invention. In the figures, like reference numerals refer to like elements. In the drawing:
• FIG. 1 is a block diagram illustrating an example of a re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 2 is a system diagram showing the flow of the hardware design, embedded software, and PCB design of the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 3 is a flow diagram of the overall re-programmable digital platform design method in accordance with one embodiment of the present invention;
• FIG. 4 illustrates designing with pre-synthesized symbolic components within a Schematic Editor included in the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 5 is a diagram illustrating employing a pre-synthesized processor core component in an FPGA design using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 6 shows pre-synthesized EDIF models allowing for vendor-independent design using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 7 is a diagram illustrating generic model linking in accordance with one embodiment of the re-programmable digital platform design system of the present invention;
• FIG. 8 illustrates specifying a target device in a constraint file using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 9 illustrates locating the correct model to use for a specific target FPGA device using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 10 illustrates a process flow when processing a captured FPGA design using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 11 illustrates that a suitable Project-Configuration combination is required to access the Process Flow when using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 12 illustrates adding the relevant constraint file to the required configuration using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 13 illustrates detecting the required Project-Configuration combination giving access to the Process Flow for the device when using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 14 illustrates a Build stage of the Process Flow for Xilinx (left) and Altera (right) devices using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 15 illustrates summarizing resource usage for a chosen device using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 16 illustrates an exemplary constraint file using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 17 illustrates a Configuration Manager using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 18 illustrates diagrams of three configurations and their constraint files using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 19 illustrates a design targeting a Xilinx Spartan2e 300 device on a NanoBoard using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 20 illustrates a design targeting an Altera Cyclone EPC12 device on a NanoBoard using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 21 illustrates a design targeting aXilinx Spartan2 100 device on a User Board using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 22 is a block diagram of a NanoBoard configuration preferably included in the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 23 illustrates accessing information for multiplexed JTAG chains over a single JTAG link when using the re-programmable digital platform design system in accordance with one embodiment of the present invention;
• FIG. 24 is a block diagram of daisy-chaining multiple development boards preferably included in the re-programmable digital platform design system in accordance with one embodiment of the present invention; and
• FIG. 25 is a block diagram showing FPGA Daughter-Board-to-Flash-RAM communications preferably included in the re-programmable digital platform design system in accordance with one embodiment of the present invention.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
• The present invention is particularly applicable to a software-and-hardware-based re-programmable digital platform design system, and it is in this context that the various embodiments of the present invention will be described. It will be appreciated, however, that the re-programmable digital platform design system and method in accordance with the present invention have greater utility, since they may be implemented in different proportions of software and hardware or may incorporate other modules or functionality not described herein.
• FIG. 1 is a block diagram illustrating an example of a re-programmable digitalplatform design system10 in accordance with one embodiment of the present invention implemented on a personal computer (PC)12. In particular, thePC12 may include adisplay unit14, which may be a cathode ray tube (CRT), a liquid crystal display, or the like; aprocessing unit16; and one or more input/output devices18 that permit a user to interact with the software application being executed by the PC. In the illustrated example, the input/output devices18 may include akeyboard20 and amouse22, but may also include other peripheral devices, such as printers, scanners, and the like. Theprocessing unit16 may further include a central processing unit (CPU)24, apersistent storage device26, such as a hard disk, a tape drive, an optical disk system, a removable disk system, or the like, and amemory28. TheCPU24 may control thepersistent storage device26 andmemory28. Typically, a software application may be permanently stored in thepersistent storage device26 and then may be loaded into thememory28 when the software application is to be executed by theCPU24. In the example shown, thememory28 may contain a re-programmable digitalplatform design tool30. The re-programmable digitalplatform design tool30 may be implemented as one or more software modules that are executed by theCPU24.
• In accordance with the present invention, the re-programmable digitalplatform design system10 may also be implemented using hardware and may be implemented on different types of computer systems, such as client/server systems, Web servers, mainframe computers, workstations, and the like. Now, more details of an exemplary implementation of the re-programmable digitalplatform design system10 in software and hardware will be described.
• Current digital circuit design tools are primarily HDL-based and not well integrated with embedded software tools. One embodiment of the re-programmable digitalplatform design system10 in accordance with the present invention is leveraged from “Board-on-Chip” technology that consists of a design technology that brings together software and hardware design processes in a single, integrated design environment to enable users to employ board-level methodologies to design and implement embedded systems on FPGAs. This technology enables users to employ board-level design methodologies to both design and implement an entire digital system, including embedded microprocessor cores and the software that runs on them, onto an FPGA. The “Board-on-Chip” design technology enables users to employ familiar design methodologies to harness the power of FPGAs as a design platform and provides users the power to transition board designs directly into FPGAs. The re-programmable digitalplatform design system10 and method in accordance with the present invention enable rapid implementation and testing of complete digital systems on FPGAs without the need for HDL source files or extensive synthesis and verification.
• In order to facilitate “Board-on-Chip” design, technology from a full range of EDA and embedded products has been brought together. Preferably, nVisage multi-dimensional design capture and TASKING embedded software technologies (including Viper multi-core compiler and debugger technologies) have been combined on the Design Explorer (DXP) platform available from Altium Limited to provide a single, integrated hardware design and software development environment. This environment enables:
• • Mixed schematic/RTL-level HDL design capture;
  • Integrated software development;
  • Processor core packs that combine pre-synthesized processor cores with matching compiler, simulator, and debugger;
  • Schematic component libraries containing a range of pre-synthesized components including common peripheral devices, generic logic, and a range of communication and interface components;
  • Primitives and macro libraries for all Xilinx and Altera device families;
  • Virtual instruments, such as logic analyzers and frequency counters, that can be built into the design for test purposes; and
  • A “Board-on-Chip” development board loaded with an FPGA that functions as a breadboard and allows implementation of the design directly from the user'sPC12 onto the FPGA.
    The “Board-on-Chip” technology provides a fully integrated and self-contained system for FPGA-based design that is accessible to users.
• Generally, software tends to remain fluid throughout the development process because it is easy to update and change. Hardware, on the other hand, tends to be locked down early in the process, because of time and cost involved in each hardware iteration and the fact that the board needs to be available in order to finalize software development.
• One of the most significant benefits of “Board-on-Chip” technology is that it allows a more flexible approach to partitioning the design between hardware and software. Users retain the ability to choose a hardware or software solution to any particular problem throughout the design process, thereby providing true parallel development of hardware and software. Also, the inclusion of a targeted development board allows hardware dependencies in the software to be addressed before the target hardware is finalized.
• Using the “Board-on-Chip” technology with its reconfigurable FPGA-based hardware platform, hardware updates can be made as easily as software updates. This makes it possible to continue to explore different hardware options throughout the design process and allows final hardware/software partitioning decisions to be delayed until late in the development process.
• The re-programmable digitalplatform design system10 in accordance with the present invention provides a systems focus to FPGA design. The re-programmable digitalplatform design system10 provides an out-of-the-box design environment for architecting entire embedded systems on FPGAs.
• The re-programmable digitalplatform design system10 provides a comprehensive, vendor-independent solution for system-level design on an FPGA platform derived from the Board-on-Chip technology described above. The re-programmable digitalplatform design system10 integrates hardware design tools, embedded software development tools, IP-based components, virtual instrumentation, and a reconfigurable development board to allow users, even those without HDL experience, to interactively design and implement a complete embedded system inside an FPGA.
• The benefits that the re-programmable digitalplatform design system10 provides to users include parallel design of hardware and software, greater flexibility in hardware/software design partitioning, an integrated FPGA vendor-independent solution for putting entire embedded systems into FPGAs, and a “live,” interactive design environment for system-on-FPGA development and debug.
• The re-programmable digitalplatform design system10 is a complete system-on-FPGA design environment built upon “live” real-time, hands-on engineering that happens inside the physical hardware design space using an approach that maps directly to a user's existing knowledge of system-level design. The re-programmable digitalplatform design system10 environment is complete and ready to use, with design hardware, design software, and IP.
• The re-programmable digitalplatform design system10 employs proven board-level system design methodologies and retargets them for FPGA architectures. The re-programmable digitalplatform design system10 also integrates hardware design and software development within a single environment to provide a total solution to systems design. The result is a system-on-FPGA tool that allows risk-free chip-level systems integration, practical hardware/software co-design, a complete systems-level development environment for FPGA-based embedded design, and introduces an interactive design methodology referred to as LiveDesign that is accessible to users.
• At the system level, the re-programmable digitalplatform design system10 provides a schematic-based design methodology to define system connectivity. This enables graphical schematic-based capture which is more efficient for connecting functional blocks than HDLs and allows complex systems to be created quickly at the component level.
• Schematic design is preferably facilitated in the re-programmable digitalplatform design system10 by the inclusion of extensive libraries of royalty-free, pre-synthesized, pre-verified IP components, including a range of processor cores, that can be simply dropped onto the schematic and connected together to form the system hardware. This is analogous to the way designers currently work at the board level with physical, “off-the-shelf” components.
• The re-programmable digitalplatform design system10 components are processed for a variety of target FPGA architectures from multiple vendors. This allows design portability between FPGA device families and ensures a flexible vendor-independent approach to FPGA design. The re-programmable digitalplatform design system10 automatically and transparently selects the correct component model for the target architecture during system synthesis. As a result, designs can be synthesized very quickly, because the component IP does not need to be reprocessed during synthesis. The re-programmable digitalplatform design system10 component system provides a novel and secure framework for FPGA IP delivery that avoids the security problems associated with supplying IP as HDL source code.
• Along with IP components, the re-programmable digitalplatform design system10 includes a library of IP-based virtual instruments, such as logic analyzers, frequency counters/generators, and IO monitors, that can be incorporated into the design at the schematic level to facilitate system debugging. Like the IP components, the virtual instruments are supplied as pre-synthesized models that allow them to be used across FPGA target architectures. These instruments have on-screen front panels analogous to their physical counterparts to provide an intuitive way for users to examine the operation of their circuit, and to “see” inside the FPGA during the design process.
• Preferably integral to the re-programmable digitalplatform design system10 is a versatile FPGA-based development board called a NanoBoard that provides a reconfigurable platform for implementing and debugging the design. The NanoBoard is connected to the user'sPC12 and uses JTAG-based communications to both download the design to the on-board FPGA, and to interact with processor cores and instruments in the design once it has been downloaded.
• Target FPGAs are housed on plug-in daughter boards to allow easy retargeting of designs. Multiple NanoBoards can be daisy-chained together to facilitate the design of complex multi-FPGA systems, and can accommodate the inclusion of end-user boards into the system for final production PCB testing and debugging.
• To facilitate system software development, the re-programmable digitalplatform design system10 includes a complete set of software development tools for all supplied processor cores to enable integrated software development. Using the Viper reconfigurable compiler technology, the re-programmable digitalplatform design system10 provides high-quality code development and debugging that is fully integrated with the hardware development environment.
• Once the target design has been downloaded to the NanoBoard, all processors in the design can be controlled and debugged from within the re-programmable digitalplatform design system10. This enables software development to take place directly on the target hardware from early in the design cycle, supporting parallel development of hardware and software. Hardware designers can download their designs to the NanoBoard for interactive debugging during development, and software designers can develop their program code directly on the hardware beginning early in the design cycle.
• Because hardware can be updated with the same ease as the software, the re-programmable digitalplatform design system10 allows more flexible design partitioning. Implementing the design on the NanoBoard means a physical prototype is not required to support completion of the debug process, so that the need to finalize hardware can be delayed until later in the development cycle.
• The re-programmable digitalplatform design system10 interactive system design environment and ability to directly connect designers and developers with their designs allows users to adopt a very hands-on approach to the development process. This provides a so-called LiveDesign design methodology.
• Unlike conventional electronics design flows, LiveDesign-enabled tools represent a new methodology for hardware and software system process flows. The LiveDesign electronics development methodology supports real-time, on-the-fly design and debug of a physical circuit with benefits for design speed, flow, quality, and cost.
• The LiveDesign methodology also enables system implementation for debug purposes, minimizing the reliance on time-consuming system-level simulation required by other design flows. There is little time or cost penalty involved in multiple design iterations, enabling users to try different design solutions without the need to physically manufacture prototype boards.
• Considered in more detail, the re-programmable digital platform design system and method in accordance with the various embodiments of the present invention bridge the gap between what has previously been a development and prototyping platform used during the design phase of an EDS project, and commercially viable re-programmable product platforms to replace non-programmable platforms, such as discrete processors and ASICS. The re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention provide for packaging, certification, and secure delivery of IP components.
• The approach employed by the re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention is based on a process of translating the required design process back into models which are familiar and intuitively manageable by most users. This allows users of the re-programmable digitalplatform design system10 and method to design as if they are using discrete components, such as processors and peripheral devices, in order to build complex systems with schematic symbology used traditionally with EDS.
• The design environment of the re-programmable digitalplatform design system10 and method enables a user to design, implement, and debug a digital design in an FPGA. The design is captured as a schematic, or using a mixture of schematic and RTL-level HDL. The embedded software is written in a coding-aware editor, ready for compilation and download.
• FIG. 2 is a block diagram showing the flow of the hardware design, embedded software, and PCB design of the re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention. As shown inFIG. 2, once a hardware design is complete, it is synthesized, a process that transforms it from the capture form into a low-level gate form.
• After design synthesis, a place and route is performed, a process during which device-aware software implements the design in the target FPGA. The vendor-specific place and route software required to synthesize for the target architecture is operated by the DXP environment, which automatically manages all project and file handling aspects required to generate an FPGA program file.
• To test and debug the design, the re-programmable digitalplatform design system10 includes a NanoBoard, an implementation platform that includes an FPGA, as well as an array of general purpose peripheral components. The software communicates directly with the NanoBoard via a port on thePC12, programming the FPGA and implementing a design by a user.
• Once the design has been implemented on the NanoBoard, the design can be debugged, using virtual instruments and boundary scan pin status technology to debug the hardware, and an integrated debugger for the embedded software. Since debugging is performed from within the same environment in which the design is captured, design iterations can be carried out quickly and software/hardware solutions rapidly explored.
• FIG. 3 is a flow diagram of the overall re-programmable digital platform design method in accordance with one embodiment of the present invention. The re-programmable digital platform design method provides a graphical design environment that employs symbolic, component-level logical blocks for selection by the user. This generic logic can be quickly assembled into complex systems. The user does not need to write any RTL- or netlist-level descriptions, as these are managed automatically by the design environment. The pre-synthesis of these component-level blocks is enabled by a module that allows components to be symbolically rendered and interconnected within a supporting graphical design environment.
• FIG. 4 illustrates designing with pre-synthesized symbolic components within a Schematic Editor. The top schematic sheet of an exemplary FPGA design that has been captured using the Schematic Editor is shown inFIG. 4. The exemplary hierarchical design, for example, a simple buzzer, utilizes a variety of symbolic components, including processor components, peripheral components, and generic logic components. Being able to place and wire FPGA-ready schematic components directly within a design capture environment already familiar to users enables them to build a more complex design simply and efficiently, without the common requirement for advanced HDL knowledge and expertise.
• FIG. 5 illustrates employing a pre-synthesized processor core component in an FPGA design.FIG. 5 shows the use of a schematic processor component within the same exemplary design shown inFIG. 4. The processor itself, for example, a member of the TSK51x family of microprocessors, is located on a sub-sheet within the design.
• The re-programmable digitalplatform design system10 and method employ a generic set of FPGA macro components comprising symbolic representations of blocks of functionality that a user may add to an FPGA. These schematic components are presented to the user as FPGA-ready schematic symbols that can be instantiated into a design. FPGA-ready schematic components are like traditional PCB-ready components, except instead of the symbol being linked to a PCB footprint, each is linked to a pre-synthesized EDIF model.
• The pre-synthesized components are supplied as object code entities without having to expose the underlying RTL-level source code. The re-programmable digitalplatform design system10 includes multiple libraries providing a comprehensive set of pre-synthesized components, ranging from simple gate-level functional blocks, up through high-level hardware, such as multipliers and pulse-width modulators, and high-level functions, such as microprocessors and communications functions.
• These components can be instantiated into designs by the system user and then the whole design can be targeted to a suitable device. The re-programmable digitalplatform design system10 and method automatically manage the resources required to instantiate the design in the chosen device, by ensuring that the EDIF models specific to that device are correctly chosen and linked to the generic symbols placed when capturing the design.
• The re-programmable digitalplatform design system10 and method preferably provide model linkage. As shown inFIG. 6, models are included for the generic, FPGA-ready, vendor-independent components and are stored in a hierarchy of folders, according to vendor and device family.
• The target FPGA device can be changed at any time during the design process, and the re-programmable digitalplatform design system10 and method will re-link all of the generic component symbols to the relevant pre-synthesized EDIF models for that device, found within these folders.FIG. 7 illustrates the underlying mechanism for this linking, enabling vendor independency while enabling the user to design by placing generic schematic symbols.
• The target device specified in the associated constraint file is used to select the correct folder of EDIF models, and then the component's Library Reference is used to select the EDIF model file from within that folder. For example, consider the LCD controller component in the simple buzzer design shown inFIG. 4, which in turn is targeted to an Altera Cyclone device using the relevant constraint file shown inFIG. 8. The re-programmable digitalplatform design system10 and method automatically link the correct pre-synthesized model for that device to the generic symbol for the component placed on the schematic sheet. Because the target device is a Cyclone and belongs to the Altera stable of device families, the \Altera\Cyclone library sub-folder is searched for the model.FIG. 9 illustrates locating the correct model to use for a specific target FPGA device. The particular model to use is specified by the Library Reference for the component symbol, in this case an LCD16X2A, as shown inFIG. 9.
• In addition to supplied models, user-created pre-synthesized EDIF models are supported. These can be stored in a single user model folder or in a hierarchy of folders if the user is developing a model for multiple target devices. The search sequence for EDIF models is:

• 	$project_dir
  	$user_edif\$vendor\$family
  	$user_edif\$vendor
  	$user_edif
  	$system_edif\$vendor\$family
  	$system_edif\$vendor
  	$system_edif

• Pre-synthesized user models are developed by creating a Core project, whose EDIF output becomes the model for a user-defined component. There are a number of features to support this process, including commands to synthesize for all targets, publish the EDIF model (that is, package it with all other required EDIF models), and generate a component symbol to represent the core.
• In accordance with one embodiment of the present invention, the re-programmable digitalplatform design system10 and method automatically re-target a generic design for a selected re-programmable device. When processing a captured FPGA design, the stages involved are all carried out using a single Process Flow. Each physical device has an associated Process Flow, as shown inFIG. 10 The Process Flow consists of four distinct stages: 1) Compile, 2) Synthesize, 3) Build, and 4) Program FPGA.
• Preferably, the re-programmable digitalplatform design system10 and method include pre-processing preparation. In order to access the Process Flow for a physical device, a suitable project-configuration combination must be available. For a combination to be suitable, a configuration must have been defined for an FPGA project, and that configuration should contain a constraint file that targets the particular device.
• Suitable project-configuration combinations are automatically detected, where they exist, and are made available in a drop-down menu below the device.FIG. 11 illustrates a case in which a device in the view does not have a suitable project-configuration combination, and, hence, the Process Flow is not available.
• In this case, a constraint file targeting the CoolRunner device would have to be created and then assigned to a configuration using a Configuration Manager included in the re-programmable digitalplatform design system10, as shown inFIG. 12. The suitable project-configuration combination will then be automatically detected, giving access to the Process Flow for the device, as shown inFIG. 13.
• The re-programmable digitalplatform design system10 and method provide mapping of a synthesized design. With respect to a design that has been synthesized, each device vendor provides a set of tools that can be used to map a synthesized design into the hardware resources inside a given target device. These tools are quite different from one vendor to another, but provide the same functionality.
• The re-programmable digitalplatform design system10 and method wrap transparently around each different tool chain and provide a generic view of the Process Flow to the user. This enables users to only have to learn one process flow in order to use multiple different vendors' tool chains and allows easy movement between different devices from different vendors to easily port their designs to new target devices.
• This flow is accessed from the Build stage of the overall Process Flow, as shown for Xilinx and Altera devices inFIG. 14. The Build stage of the Process Flow is used to run the vendor place and route tools. Running the tools at this stage can verify whether or not a design will indeed fit inside the selected device. The user may also elect to run the vendor tools if he or she wants to obtain pin assignments for importing back into the relevant constraint file.
• The end result of running the Build stage is the generation of an FPGA programming file that is ultimately used to program the target device with the design. There are preferably five main stages to the Build process:
• • Translate Design—uses the top-level EDIF netlist and synthesized model files, obtained from the synthesis stage of the Process Flow, to create a file in Native Generic Database (NGD) format in the case of Xilinx target devices
  • Map Design To FPGA—maps the design to FPGA primitives
  • Place and Route—takes the low-level description of the design from the mapping stage and determines how to place the required logic inside the FPGA, and, once arranged, the required interconnections are routed
  • Timing Analysis—performs a timing analysis of the design, in accordance with any timing constraints that have been defined, and, if there are no specified constraints, default enumeration is used
  • Make Bit File—generates a programming file that is required for downloading the design to the target device.
    When targeting a Xilinx device, an additional stage is available, namely, Make PROM File. This stage is used when the user wants to generate a configuration file for subsequent download to a Xilinx configuration device on a production board.
• After the Build stage has completed, the Results Summary dialog appears, as shown inFIG. 15. This dialog provides summary information with respect to resource usage within the target device.
• Additionally, the re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention automatically manage device-specific design constraints. The design captured in the schematics and RTL-level HDL defines the behavior of the circuit. To map this design into an FPGA that is part of a larger PCB design requires the addition of all the necessary implementation information. This information may include:
• • the target PCB project;
  • the target FPGA;
  • FPGA net-to-physical device pin assignments;
  • pin configuration information, such as output voltage settings;
  • design timing constraints, such as allocating a specific net to a special function FPGA pin; and
  • FPGA place and route constraints.
• The re-programmable digitalplatform design system10 and method provide a generic constraint model in which this information can be specified. By storing the implementation data separate from the source design documents, the same design can be easily re-targeted to a different physical FPGA device, allowing true design portability.
• The implementation information is stored in constraint files. A constraint file is a set of constraint records, in which each record or constraint group defines one or more constraints to be applied to a particular object in the FPGA project. The FPGA-specific constraint definitions used in a constraint file are those recognized by vendor tools, enabling transparent mapping of constraint file information to the vendor-specific constraint system for implementation.
• FPGA designs are captured in the DXP-based design environment as schematics and RTL-level HDL files. Implementation information, such as device pin allocations and pin electrical properties, are not stored in these source files; they are stored in separate constraint files.
• Constraints can be divided over a number of constraint files, and a project can include a number of different constraint files that are used to target the design to different implementations. To allow a user to group constraint files and to move easily from one implementation to another, the user defines configurations. A configuration is simply a named collection of constraint files.
• Constraints are normally defined in constraint files using a constraint editor. To add a new constraint file to a project, a user can right-click on the project name, then from an Add New to Project submenu select Constraint File. Constraints can be defined by typing them in, or by adding them using menu options available in a constraint file editor Design menu.
• Constraint files are mapped to an FPGA design by defining a Configuration. The user can do so by right-clicking on the project file name in the panel and selecting Configuration Manager. Multiple constraint files can be enabled in a configuration, enabling the user to separate constraints by their type. For example, a user can add design-specific constraints, such as specifying a net to be a clock, in one constraint file, and implementation type constraints, such as pin allocations, in a second constraint file. When an FPGA design is implemented in a device, the user selects a project/configuration combination, allowing the design to be mapped to its implementation. By defining multiple configurations, the user can easily re-target a design from one device to another.
• Typically, constraint information is divided across multiple constraint files in accordance with the particular class of constraint being defined. For example, the classes may include:
• • Constraints specific to the device and board—concerning connection of the FPGA device to the outside world;
  • Constraints specific to a certain device—concerning specifications for how to use internal features of the particular device. A good example would be place and route controls for FPGA tools which are very device-specific, but do not have any impact on the way the device is connected to the outside world; and
  • Constraints specific to a project or design—concerning requirements which are associated with the logic of the design, as well as constraints on its timing.
• To ensure the most portable FPGA design, the most logical approach is to break the constraints into these three classes and store these in separate constraint files, as shown inFIG. 16.FIG. 14 illustrates an exemplary constraint file, with two clock nets constrained by the FPGA_CLOCK=TRUE constraint. Note that the pin allocations have not yet been defined.
• Preferably, constraint definitions include: 1) Record to define the purpose of entry in a constraint file; 2) TargetKind to define what a record in the constraint file applies to; 3) TargetId to provide the name of the object being targeted by the constraint record; and 4) Id, which is a FileHeader constraint to specify the type and version of constraint file.
• FPGA-specific constraints are used in conjunction with the device vendor's component specifications to ensure that a particular constraint is supported in that device. These include the following FPGA-specific constraints:
• FPGA_CLOCK states that the top level port of the target technology should use a high speed resource.
• FPGA_CLOCK_DUTY_CYCLE sets the duty cycle of the clock.
• FPGA_CLOCK_FREQUENCY sets the desired frequency of the clock, and the place and route tool will attempt to achieve this frequency.
• FPGA_CLOCK_PIN states that the top level port is connected to a reserved clock pin on the target device.
• FPGA_DEVICE specifies the target device.
• FPGA_DRIVE specifies the current strength of a pin in the target device and may be applied to ports in the top-level entity.
• FPGA_GLOBAL states that the signal should be kept, and use a high speed resource.
• FPGA_INHIBIT_BUFFER states that no input/output (I/O) buffers are inserted at the pin.
• FPGA_IOSTANDARD specifies the pin I/O standard in the target device, and may be applied to ports in the top-level entity.
• FPGA_KEEP prevents a particular signal from being optimized.
• FPGA_NOCLOCK prevents the synthesizer from placing a clock buffer. The synthesizer has the ability to infer clock buffers and add them automatically.
• FPGA_PCI_CLAMP enables Peripheral Component Interconnect (PCI) compatibility for a pin.
• FPGA_PINNUM specifies the pin-out in the target device, and may be applied to ports in the top-level entity.
• FPGA_PULLDOWN adds a pulldown resistor to an I/O port.
• FPGA_PULLUP adds a pullup resistor to an I/O port.
• FPGA_RESERVE_PIN ensures that the place and route tools do not assign this particular pin to a port of the design.
• FPGA_SLEW specifies the slew standard in the target device, and may be applied to ports in the top-level entity.
  The user can also invoke the options in the Design>>Import Pin File menu in the constraint file editor to import constraints defined by the place and route software.
• The re-programmable digitalplatform design system10 includes a Configuration Manager, allowing the user to determine which constraint files are to be associated with a particular target FPGA device, as shown inFIG. 17.FIG. 15 illustrates a Configuration Manager showing three configurations, each with one constraint file assigned.
• For example, consider an FPGA project that is targeted to:
• • a Xilinx Spartan-XC2S300E QFP208 on a NanoBoard,
  • an Altera Cyclone QFP240 on a NanoBoard, and
  • a Xilinx Spartan-XC2S100E QFP144 on the user's own board design.
    This project requires three configurations, one for each target device. Assuming good design practice from a portability perspective, this would theoretically require seven constraint files, separate constraint files to control the pin-outs for each of the three devices in their target boards, separate constraint files to control any internal place and route constraints for each of the three target devices, and one constraint file for logical design constraints. If there were no place and route constraints, then there are only four constraint files. Each configuration would then include three constraint files, two which are specific to the configuration and one which is common to all configurations.
• If there were constraints that were common to the two different Xilinx devices, which are internally very similar, then it may be beneficial to create a constraint file for these. In this case, the extra constraint file would be added to two of the configurations.
• Diagrams of these three configurations and their constraint files are illustrated inFIG. 18.FIG. 18 shows the relationships between the constraint files and the configurations. As shown inFIG. 18, a single set of schematic and RTL-level HDL source documents is targeted to three different implementations by different sets of constraint files. The actual setups for each of the three configurations are shown inFIGS. 19-21.
• Each line in a constraint file is referred to as a constraint group, since it applies one or more constraint requirements to an entity in the design. The following examples show a variety of different constraint groups.
EXAMPLE 1Record=Constraint | TargetKind=Part | TargetId=XC2S300E-6PQ208C
• This constraint targets the part (or FPGA component), in this case specifying the type as XC2S300E-6PQ208C. Note that this device must be supported by selecting the Design>>Add/Modify Constraint>>Part menu item to display the Choose Physical Device dialog. This dialog displays all devices supported by the re-programmable digitalplatform design system10. This constraint group should only appear once in any configuration.
EXAMPLE 2

• 	Record=Constraint | TargetKind=Port | TargetId=VGA_VSYN |
  	FPGA_PINNUM=P134

  
  This constraint group targets the net connected to port VGA_VSYN, specifying that it is connected to FPGA pin number P134.
EXAMPLE 3

• Record=Constraint | TargetKind=Port | TargetId=LCD_E |
  FPGA_PINNUM=P82 | FPGA_IOSTANDARD=LVTTL33 | FPGA_SLEW=SLOW |
  FPGA_DRIVE=12mA

  
  This constraint group targets the net connected to the port LCD_E, specifying that it is connected to FPGA pin number P82, have an I/O standard of LVTTL33, a slew setting of slow, and a drive current of 12 mA.
EXAMPLE 4

• 	Record=Constraint | TargetKind=Port | TargetId=VGA_R[1..0] |
  	FPGA_PINNUM=P145,P141

  
  This constraint group targets both the nets in the bus connected to the port VGA_R[1.0], specifying that thebus element1 connect to pin P145, andbus element0 connect to pin P141.
• The re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention also preferably provide a development platform for interactive, re-targetable hardware/software co-design of programmable device hosted systems. This development platform, referred to as a NanoBoard, allows a single development platform to provide development resources for an unlimited number of target devices, such as FPGAs, CPLDs, or embedded microprocessors, from one or more different semiconductor manufacturers. The NanoBoard contains a controller that manages communications with the host-based development software, the daisy-chaining of multiple NanoBoards, connections to user-designed boards, and interfaces to physical resources on the NanoBoard.
• The NanoBoard controller employs a “NanoTalk” protocol to communicate with the PC-based software and to communicate with other NanoBoards in the daisy-chain. Each NanoBoard also contains a “standard” socket for installing various daughter boards, each of which houses a physical device that is to be used for development. These daughter boards generate an identification code that is read by the controller, which can then program the device at power-up.FIG. 22 is a block diagram of a NanoBoard, showing the interface between the on-board controller (NanoTalk controller) and both thehost PC12 and the daughter board plug-in.
• The re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention further preferably provide an interactive design environment that automates support for multiple development boards. Each NanoBoard has a master and slave connector, allowing an unlimited number of NanoBoards to be daisy-chained together. When another board is plugged into the chain, it is automatically recognized, and all of its resources become available to the re-programmable digitalplatform design system10.
• The chain of development boards may consist of multiple NanoBoards, as well as user or third party development boards, attached to the chain through the use of dedicated user board headers, located on each NanoBoard. The re-programmable digitalplatform design system10 employs a single pin on a header (either the NanoTalk Master or NanoTalk Slave headers when chaining additional NanoBoards, or User Board A/User Board B headers when attaching user boards) to detect and allow another board with JTAG devices to be automatically brought into the main JTAG chain.
• Detection and resource management is controlled by the NanoBoard controller on the Master board which, upon detecting the presence of a new Slave board brings all of the hard devices for that board into the main Hard JTAG chain and all of the soft devices into the main Soft JTAG chain. A protocol for multiplexed JTAG/SPI channels is used, in which a single JTAG interface between thePC12 and the development platform controls multiple JTAG and SPI chains, as shown inFIG. 23.
• FIG. 24 illustrates two NanoBoards daisy-chained together, with a third party development board attached to the first NanoBoard in the chain. InFIG. 24, only TDI and TDO JTAG signals have been shown for simplicity. Multiplexing of JTAG and SPI chains into a single JTAG link between thePC12 and the first NanoBoard in the daisy-chain is handled by the NanoBoard controller of that first board.
• All routing of the main Hard and Soft JTAG chains is carried out within the NanoBoard controller, with each chain routed in accordance with detected boards, such as additional NanoBoards, user or third party boards, and daughter boards. The NanoBoard controller on each board can be accessed from thePC12 to control the individual resources located on each board.
• The re-programmable digitalplatform design system10 and method in accordance with one embodiment of the present invention also preferably provide a generic boot system capable of supporting multiple programmable device architectures. The re-programmable digitalplatform design system10 incorporates a generic boot system providing support to program, at power-up, a multitude of different target devices. This bootstrapping functionality is facilitated using the NanoBoard controller and a generic low-cost serial flash random access memory (RAM) device.
• The flash RAM is an SPI-compatible device. Access to this RAM by the physical device on the daughter board is made through the SPI controller, which resides within the NanoBoard controller on the NanoBoard, as shown inFIG. 25.
• The SPI controller serves the role of SPI Bus Master, determining what accesses the SPI Bus (NanoBoard controller, daughter board, or host computer (via the parallel port)) and which of the SPI slave devices (embedded flash RAM, FPGA boot flash RAM, or NanoBoard system clock) is selected for communications. Preferably, the re-programmable digitalplatform design system10 in accordance with one embodiment of the present invention comprises a programmable clock generator to facilitate the live testing of a design in the target device at different operating frequencies by allowing thehost PC12 to dynamically select a clock frequency from a range of allowable frequencies. Thus, thehost PC12 may be employed to vary the FPGA clock frequency.
• The daughter boards generate an identification code which is used by the NanoBoard controller to identify the device. At power-up, the following occurs:
• • the NanoBoard controller identifies the physical device on the currently plugged-in daughter board
  • the SPI controller is configured for communications between the daughter board and the flash RAM device
  • the data stream is read from the flash RAM, and the physical device on the daughter board is subsequently programmed using the programming algorithm that is appropriate for that target device.
• To assist with interchanging FPGA daughter boards with different power rail requirements, the re-programmable digitalplatform design system10 in accordance with one embodiment of the present invention preferably comprises a programmable voltage regulator. The target voltage is determined by a divider network located on the daughter board and connected to the NanoBoard of the re-programmable digitalplatform design system10 via a designated pin. This configuration enables the size and cost of the daughter boards to be reduced while increasing performance.
• The following is an example of textual source code produced by one embodiment of the design system and method for re-programmable digital platforms in accordance with the present invention:
• The re-programmable digitalplatform design system10 and method in accordance with the various embodiments of the present invention have wide applicability across the electronics industry, and are beneficial in many industries, where product value is relatively high, but market size does not allow conventional ASIC development. The re-programmable digitalplatform design system10 and method provide immediate benefits to:
• • Engineers who routinely develop digital systems using off-the-shelf silicon devices, because the re-programmable digitalplatform design system10 and method provide an intuitive approach for migrating board-level systems into device-level designs;
  • FPGA designers looking for an easier way to embed soft processor cores into their designs, because the re-programmable digitalplatform design system10 and method provide a reconfigurable development environment, the tools, and EP to support complete system implementation;
  • Hardware and software engineers designing low- to medium-complexity embedded microcontroller applications, because the re-programmable digitalplatform design system10 and method integrate hardware and software development flows from the first line of code through board-level implementation.
• The re-programmable digitalplatform design system10 and method represent a new way of approaching digital design. In particular, the re-programmable digitalplatform design system10 and method provide:
• • Shorter embedded systems development cycles by enabling parallel design of hardware and software;
  • Greater flexibility in hardware/software design partitioning;
  • An integrated solution for putting entire embedded systems into FPGAs;
  • The benefits of chip-level integration to all users, irrespective of their HDL knowledge and expertise;
  • An ASIC alternative to high-cost factory ASIC implementation;
  • An FPGA vendor-independent solution to systems design; and
  • A “live,”, interactive environment for system-on-FPGA development and debug.
• While the foregoing description has been with reference to particular embodiments of the present invention, it will be appreciated by those skilled in the art that changes in these embodiments may be made without departing from the principles and spirit of the invention. Accordingly, the scope of the present invention can only be ascertained with reference to the appended claims.

Claims (20)

1. A method for digital circuit design, comprising the steps of:

placing and wiring schematic components directly within a design capture environment familiar to users to enable them to build a design without the common requirement for advanced HDL knowledge and expertise;

instantiating the components into a design by the user; and

targeting the design to a re-programmable device;

thereby enabling the user to design by placing generic schematic symbols to produce a design.

2. The method ofclaim 1, further comprising the step of:

providing pre-synthesized components as object code entities without having to expose the underlying RTL-level source code.

3. The method ofclaim 1, further comprising the step of:

providing multiple libraries comprising a comprehensive set of pre-synthesized components, ranging from simple gate-level functional blocks, up through high-level hardware, and high-level functions.

4. The method ofclaim 1, further comprising the steps of:

providing models for generic, vendor-independent components; and

storing the models in a hierarchy of folders according to vendor and device family.

5. The method ofclaim 4, further comprising for the step of:

enabling a target re-programmable device to be changed at any time during in the design; and

re-linking all of the generic component symbols to the relevant pre-synthesized EDIF models for the device, found within the folders;

thereby enabling vendor independency.

6. The method ofclaim 4, further comprising the steps of:

providing user-created pre-synthesized EDIF models; and

storing the user-created models in one of a single user model folder or in a hierarchy of folders if the user is developing a model for multiple target devices.

7. The method ofclaim 1, further comprising the step of:

automatically re-targeting a generic design for a selected re-programmable device.

8. The method ofclaim 1, wherein the re-programmable device is an FPGA, further comprising the steps of:

using a single Process Flow when processing a captured FPGA design, the Process Flow consisting of four stages: 1) Compile, 2) Synthesize, 3) Build, and 4) Program FPGA.

9. The method ofclaim 1, further comprising the step of:

automatically managing device-specific design constraints associated with implementation data.

10. The method ofclaim 9, further comprising the step of:

storing the implementation data separate from source design documents;

thereby easily re-targeting the same design to a different physical FPGA device, allowing design portability.

11. The method ofclaim 1, further comprising the step of:

providing a development platform for interactive, re-targetable hardware/software co-design of a re-programmable device;

thereby allowing a single development platform to provide development resources for an unlimited number of target devices from one or more different semiconductor manufacturers.

12. The method ofclaim 11 wherein the development platform comprises a standard socket for installing various daughter boards, each of which houses a physical device that is to be used for development.

13. The method ofclaim 12 wherein each daughter board generates an identification code that is read by a controller, which can then program the physical device at power-up.

14. The method ofclaim 11, further comprising the steps of:

interconnecting a plurality of development platforms; and

employing a single pin on a header to detect and allow another development platform with JTAG devices to be automatically brought into a main JTAG chain.

15. The method ofclaim 1, further comprising the step of:

providing a generic boot system capable of supporting multiple re-programmable device architectures.

16. The method ofclaim 15 wherein the generic boot system provides support to program, at power-up, multiple re-programmable device architectures from different vendors.

17. The method ofclaim 15 wherein the boot system comprises a controller and a generic low-cost serial flash RAM device, and further comprising the step of storing in the flash RAM configuration data, wherein a type of re-programmable device from multiple vendors installed on a daughter board is read via an identification code from the daughter board and based on the identification, the configuration data from the flash RAM is used to select a correct programming algorithm to program the re-programmable device.

18. The method ofclaim 14, further comprising the step of providing a single set of wires to handle multiple multiplexed JTAG and SPI communication channels.

19. The method ofclaim 11, further comprising the step of controlling a clock frequency using of the re-programmable device using host-based development software.

20. The method ofclaim 12, further comprising the step of providing a programmable power supply on the development platform for accommodating daughter boards with different power rail requirements.

Images